CodedPaddedFL and CodedSecAgg: Straggler Mitigation and Secure Aggregation in Federated Learning
Journal article, 2023

We present two novel federated learning (FL) schemes that mitigate the effect of straggling devices by introducing redundancy on the devices' data across the network. Compared to other schemes in the literature, which deal with stragglers or device dropouts by ignoring their contribution, the proposed schemes do not suffer from the client drift problem. The first scheme, CodedPaddedFL, mitigates the effect of stragglers while retaining the privacy level of conventional FL. It combines one-time padding for user data privacy with gradient codes to yield straggler resiliency. The second scheme, CodedSecAgg, provides straggler resiliency and robustness against model inversion attacks and is based on Shamir's secret sharing. We apply CodedPaddedFL and CodedSecAgg to a classification problem. For a scenario with 120 devices, CodedPaddedFL achieves a speed-up factor of 18 for an accuracy of 95% on the MNIST dataset compared to conventional FL. Furthermore, it yields similar performance in terms of latency compared to a recently proposed scheme by Prakash et al. without the shortcoming of additional leakage of private data. CodedSecAgg outperforms the state-of-the-art secure aggregation scheme LightSecAgg by a speed-up factor of 6.6-18.7 for the MNIST dataset for an accuracy of 95%.

secure aggregation

privacy

linear regression

gradient codes

straggler mitigation

Coded distributed computing

federated learning

Author

Reent Schlegel

OHB DIGITAL SERVICES GMBH

Simula UiB

Siddhartha Kumar

Qamcom Research & Technology

Simula UiB

Eirik Rosnes

Simula UiB

Alexandre Graell I Amat

Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks

Simula UiB

IEEE Transactions on Communications

0090-6778 (ISSN) 15580857 (eISSN)

Vol. 71 4 2013-2027

Subject Categories

Telecommunications

Computational Mathematics

Computer Science

DOI

10.1109/TCOMM.2023.3244243

More information

Latest update

6/2/2023 1